readme.txt

支持向量机作为统计学习理论的实现方法

% Statistical Pattern Recognition Toolbox, Vojtech Franc, Vaclav Hlavac
% (c) Czech Technical University Prague, http://cmp.felk.cvut.cz
% Written Vojtech Franc (diploma thesis) 26.02.2000
% Modifications
% 24. 6.00 V. Hlavac Text polished.
=====================================================================

0. Contents
===========
0. Contents
1. Introduction
2. Requirements
3. Installation
4. Quick Start
5. How to start ?

1. Introduction
====================
This toolbox implements a selection of major statistical pattern
methods as described in the monograph Schlesinger M.I., Hlavac
V. Ten lectures from the statistical and syntactic pattern
recognition (in Czech), Czech Technical University Publishing
House, September 1999 (further referred as SH10). English
version should appear in Kluwer Academic Publishers. The basic
idea of the toolbox was sketched by M.I. Schlesinger and V.
Hlavac in June 1999. The design of the toolbox and
implementation is by V. Franc who did it as his diploma work at
the Czech Technical University Praha, Czech Republic (defended
February 2000). V. Franc continues developing the toolbox as he
is the PhD student of the Center for Machine Perception, cf.
http://cmp.felk.cvut.cz

The aim of the toolbox is to
(a) Provide tools for experimenting with various statistical
pattern recognition methods.
(b) Aid the reader of the above mentioned rather theoretical
book the practical skills.
(c) Provides the teachers of pattern recognition courses the
tool they can use with their students in laboratory exercises.

This software can be used freely for academic, non-profit
purposes. If you intend to use it for commercial development,
please, contact us.

2. Requirements
================
- Matlab, version 5.2 and higher.
- Optimization Toolbox (MathWorks).
- Statistic Toolbox (MathWorks).


3. Installation
================
The "stprtool" directory is a root directory of the statistical
pattern recognition toolbox.

- Create a directory of your choice and copy the toolbox there.
- The path to toolbox directories has to be set before using
  the toolbox. This can be done automatically each time Matlab
  is initiated or manually. We describe the first option here
  and the second is described in the next section.
- Windows: Copy file "\stprtool\setpath.m" to your "..\matlab\toolbox\local\"
  directory.
  Linux:  Copy file "\stprtool\setpath.m" to "\your_home\matlab\"
- Add line
        setpath('path');
 to your "startup.m" file. The argument path is the  name of the toolbox
 root directory. For example, if the toolbox is located in your root
 directory on disk c then use
        setpath('c:\stprtool');

note:
 The "startup.m" file is located in 
    Windows: "..\matlab\toolbox\local\" directory, 
    Linux: "\your_home\matlab\" directory
 if it does not exist you have to create it.


4. Start toolbox with manual set of the path
===============
- Run Matlab
- Change directory to the toolbox root. Use Matlab command "cd",
  for example
        cd c:\stprtool
- Type
        setpath


5. How to start ?
====================

After the toolbox is installed, run Matlab and you can start
using the toolbox. At first, try to type
   help stprtool

and you will get a list of available demo programs. You can also 
read html-documentation in "/stprtool/doc/html/".

If you have any comments or suggestions then send
me an email at [e-mail]